Background and objectives: Hand grip strength (HGS) has emerged as a predictor of the nutritional status. However, many factors may modify the malnutrition-HGS association. This study explored the nutritional assessment value and determinants of HGS in patients hospitalized with cancer.
Methods and study design: In this multicenter, retrospective, observational study (11,314 patients), the Receiver operator characteristic curve was used to observe HGS and nutritional status sensitivity/specificity. Sex; age; height; weight; mid-upper arm circumference (MAMC); Patient-Generated Subjective Global Assessment (PG-SGA) score; Karnofsky score; physical function (PF) domain; cognitive function (CF) domain; global health and quality of life (QL) domain of EORTC QLQ-C30 (a quality of life instrument designed by the European Organization for Research and Treatment of Cancer); and albumin, prealbumin, and hemoglobin levels were included in a Stepwise analysis model to identify the factors influencing HGS.
Results: HGS showed a very low diagnostic value and accuracy for identifying severe malnourishment (area under the curve, 0.615-0.640; p<0.01). HGS positively correlated with sex; height; weight; MAMC; Karnofsky score; QL, PF, and CF domains; and hemoglobin and prealbumin levels (Beta= 0.02-0.42, p<=0.05), and negatively with age (Beta=-0.19, p<0.01). However, the PG-SGA score was excluded because of its very limited contribution to HGS variability.
Conclusions: HGS is a mutifactorial index. The use of HGS cutoff values to identify malnutrition is markedly challenging. Thus, HGS may be of limited use as a predictor of nutritional status.